Revealing the Role of Corporate Social Responsibility, Service Quality, and Perceived Value in Determining Customer Loyalty: A Meta-Analysis Study

Abstract

This study aims to unveil the role of corporate social responsibility (CSR), service quality, and perceived value in shaping customer loyalty. A meta-analysis was conducted using 123 primary studies published between 2004 and 2024, with effect sizes extracted and analyzed to determine the strength and consistency of these relationships across diverse contexts. The results reveal that CSR positively influences both customer satisfaction and customer loyalty. Additionally, service quality and perceived value play crucial roles in fostering customer loyalty through customer satisfaction. A higher level of service quality and perceived value enhances satisfaction, ultimately leading to greater customer loyalty. The findings highlight that customer satisfaction is a key driver of customer loyalty, as satisfied customers are significantly more likely to remain loyal. Demographic characteristics such as gender, age, and education background significantly moderate the relationships among CSR, perceived value, customer satisfaction, and customer loyalty. Specifically, females tend to be more sensitive to CSR initiatives than males, older consumers show higher loyalty than younger ones, and those with higher educational levels exhibit comparatively lower loyalty. These insights enrich the existing literature and provide practical guidance for marketers and policymakers in designing targeted strategies to strengthen customer loyalty.

1. Introduction

In today’s competitive economy, retaining existing consumers is as critical, if not more so, as obtaining new ones [1]. Customer loyalty is an enduring determination to periodically repurchase a favored product or service in the future [2]. Customer loyalty is vital for an organization, since it boosts profitability, improves sales success, and enables sustainable growth [3]. Customer loyalty is vital to a brand’s long-term competitive edge over rivals, and a key target in the marketing sphere [4]. Previous research has examined a wide range of factors influencing customer loyalty across various sectors, including retail, banking, telecommunications, hospitality, and tourism [5,6,7,8,9,10]. However, a comprehensive literature analysis to impose trends, conflicts, or other intriguing linkages between corporate social responsibility, service quality, perceived value, customer satisfaction, and customer loyalty has yet to be conducted.

Corporate social responsibility (CSR) enhances a company’s image and reputation, positively impacting business viability and performance. It boosts employee motivation, retention, and recruitment by improving the work environment. Additionally, CSR fosters productivity by encouraging alternative production processes and attracting investors focused on sustainability. It mitigates or controls CSR-related risks, such as avoiding adverse press coverage [11]. Previous studies investigated the direct and indirect effects of CSR on customer satisfaction for developing customer loyalty [7,12,13,14]. These studies highlighted that CSR promotes customer satisfaction and customer loyalty, which encourages managers of firms to invest in CSR activities; this may be considered as a service marketing approach, providing a business with the opportunity to accomplish its sustainability objective. However, the underlying mechanism by which CSR influences customer satisfaction in shaping customer loyalty remains insufficiently explored in the existing literature. This study aims to address this gap by investigating the role of CSR in fostering customer loyalty via the following research question: “How does CSR enhance customer loyalty to contribute to a company’s long-term success?”

Service quality is widely recognized as a key determinant of customer satisfaction across various contexts. It remains one of the most extensively studied topics in the marketing literature due to its well-established associations with critical outcomes such as customer satisfaction and customer loyalty [15,16,17]. Service quality aids in the development of effective customer acquisition tactics and is an indicator of client loyalty [18,19]. For that reason, enhancing service quality is essential for cultivating customer loyalty, particularly within service-driven industries. This study aims to investigate the role of service quality in shaping customer loyalty by addressing the following research question: “Why does service quality take time to effectively influence customer loyalty?”

Scholars have previously recognized perceived value as a critical component in enterprise marketing strategies and a key determinant of consumer loyalty [20,21]. They emphasized that higher levels of customer engagement increase the likelihood of perceived value, which in turn drives greater customer loyalty. Ref. [22] suggested that perceived value is the antecedent of customer satisfaction, while customer loyalty is the result of customer satisfaction. However, few studies provide perceived value insights in the service industry [23,24,25]. This study aims to fill this gap by uncovering the effect of perceived value on customer loyalty via customer satisfaction.

Although customer loyalty is critical for business success, as it fosters repeat purchases, there remains a lack of comprehensive research that synthesizes existing academic findings through robust theoretical frameworks. To address this gap, the present study conducts a meta-analysis of empirical research to systematically identify the key predictors of customer loyalty [26,27]. Additionally, traditional data analysis and interpretation methods can yield inaccurate conclusions due to methodological artifacts such as measurement errors and the dichotomization of continuous variables. Meta-analytic techniques, by contrast, offer a more accurate and reliable approach by correcting for these biases through the use of mean correlations and d-values adjusted for downward bias [28,29]. This study aims to analyze key concepts and systematically organize the scientific knowledge related to the antecedents of customer satisfaction and customer loyalty. Moreover, this study responds to the call of [5,30] by investigating the potential moderating effects of demographic characteristics (gender, age, education background) on the link between CSR, perceived value, customer satisfaction, and customer loyalty through the lens of a meta-analysis.

This study has three specific objectives: (1) to investigate the influence of CSR, service quality (SQ), and perceived value on customer loyalty through customer satisfaction; (2) to examine the effect of customer satisfaction on customer loyalty; and (3) to explore the moderating role of demographic variables (gender, age, education background) in the relationship between CSR, perceived value, customer satisfaction, and customer loyalty. This research adds to the literature on customer loyalty by employing a meta-analytic methodology in three respects: First, it scrutinizes the role of customer loyalty antecedents in shaping customer behavior via the prism of social change and the expectancy–value theory. Second, it provides a valuable contribution to the field of marketing by being one of the few studies that adopts a meta-analytic approach to examine customer loyalty-related hypotheses. Third, it is one of the first studies to employ meta-analytic approaches to examine the moderating effects of demographic variables on the relationship between CSR, perceived value (PV), customer satisfaction (CS), and customer loyalty (CL).

This paper is structured as follows: it begins with theoretical foundations, followed by developing a model and hypotheses. Next, the methodological approach is then described in detail, followed by the statistical analysis results. Finally, this paper concludes by discussing implications, limitations, future research directions, and conclusions.

2. Literature Review

2.1. Social Change and Expectancy–Value Theory

Social exchange theory (SET) is a fundamental theory of social interaction in the social sciences [31]. SET, drawn from economic exchange theory, explicates human behavior in social interactions [32]. According to SET, behaviors can be considered the outcome of cost–benefit assessments of individuals attempting to participate in society and surroundings. If individuals perceive that the benefits of a behavior outweigh its costs, they are likely to engage in it. SET thus suggests that such behaviors are often rooted in long-term relationships driven by intrinsic and extrinsic rewards [33]. CSR possesses the potential to generate social exchange interactions between an organization and its customers [34]. These interactions frequently evoke emotional responses, which play a crucial role in shaping customers’ perceptions of the overall service experience, either positively or negatively. Notably, a positive experience at one service location can influence customers to form more favorable attitudes toward the brand as a whole. This generalization of positive perception across multiple service points fosters heightened customer expectations, strengthens emotional bonds with the brand, and ultimately enhances brand loyalty and long-term engagement [35].

According to expectancy–value theory (EVT, ref. [36]), motivation to engage in a specific behavior is determined by two key components: (i) expectancy, referring to the perceived likelihood that a particular action will lead to a desired outcome; and (ii) value, reflecting the importance or desirability of that outcome to the individual. In a consumer context, individuals often form expectations about a product, including its benefits and the likely consequences of its use. This cognitive evaluation informs their motivation to act. As early research suggests, individuals tend to adopt behaviors they believe will yield positive and valued outcomes [37]. In addition, EVT posits that an individual’s desire to complete a task is influenced by both the expected outcomes and the PV associated with that task [38]. The value component (subjective task value) can be divided into attainment value, intrinsic value, utility value, and cost [39,40]. This study adopted SET and EVT to explain the route effect of CSR, PV, and SQ on CL through CS.

2.2. Hypothesis Development

2.2.1. CSR and Customer Satisfaction

Ref. [41] defined CSR as “the duty of business is to advocate those policies, to make those decisions, or to follow those lines of action which are advantageous in terms of the goals and values of the community”. CSR can be defined as a company’s commitment to minimizing or eliminating any detrimental consequences while maximizing its long-term beneficial effects on the community [42]. The UNIDO (United Nations Industrial Development Organization) defines CSR as a management concept in which businesses incorporate social and environmental issues into their business activities and contacts with stakeholders. The Commission of the European Communities defined CSR as “a notion whereby enterprises integrate social and environmental issues in their business operation and in how they interact with their stakeholders voluntarily” [43].

Previous studies discovered that CSR has a significant impact on CS [8,13,44,45]. Ref. [5] revealed that high levels of corporate ethical standards are found to induce consumers to assume that the company is devoted to its CSR activities. The organization’s activities for societal, community, and environmental welfare play a significant role in shaping CS [7]. Ref. [8] conducted a study involving 283 respondents to examine the relationship between CSR and CL within the Islamic banking sector. This study revealed that CSR serves as a key driver of CS. CSR initiatives strengthen the brand image and PV of products, thereby fostering higher levels of CS [14]. Hence, we state the following hypothesis:

H1. 

CSR is positively related to customer satisfaction.

2.2.2. Service Quality and Customer Satisfaction

Ref. [46] assume that SQ is a client judgment made by comparing expectations to the delivery reality in each service facet (attribute). Ref. [47] defined SQ as “the extent of discrepancy that links customers’ expectations and perceptions”. SQ displays customers’ subjective judgment of the whole service and its features [48]. SQ is essential for increasing profitability and is at the core of the organization’s strategy to obtain a competitive edge [49].

Prior studies have demonstrated that SQ plays a pivotal role in shaping CS across various sectors, including banking, retail, and telecommunications [15,16,19,50,51]. SQ significantly influences customer evaluations of a company, as higher levels of SQ are associated with increased CS [52,53]. Ref. [54] found that adherence to quality standards in service delivery significantly enhances CS. In the tourism sector, most studies focusing on restaurants found that SQ has a positive influence on CS [30,55,56,57]. The dimensions of SQ (i.e., food quality, facility comfort, and cleanliness) significantly influence customers’ perceptions of SQ, thereby contributing to increased CS [58]. Ref. [59] emphasized that delivering outstanding SQ can significantly enhance CS that fosters CL and encourages long-term customer retention. SQ is widely recognized as a strong predictor of CS [60]. Hence, we formulate the following hypothesis:

H2. 

Service quality is positively related to customer satisfaction.

2.2.3. Percieved Value and Customer Satisfaction

PV refers to a customer’s overall appraisal of a product’s utility based on perceived benefits. Value represents the exchange between what is given and what is acquired [61]. PV is referred to as a set of benefits and compromises that the customer receives and gives, with the most typical ratio being quality to price [62]. Ref. [63] stated that PV is an indicator of the comprehensive effectiveness of a service as experienced by a person, taking into account the individual’s devotion to using the surface and the benefits obtained from it. Ref. [64] found that PV plays a crucial role in driving CS through value-added mobile services. Ref. [65] investigated the dimensions of SQ and its impact on CS in the banking sector. They found that PV has a significantly positive effect on CS, suggesting that higher levels of PV contribute to greater customer satisfaction. Ref. [25] discovered that the PV of personnel’s delivery service was a significant indicator of CS. Ref. [9] discovered that PV has both direct and indirect effects on CL via CS. They stated that PV reflects the evaluation/appraisal stage, CS is the reactions to an emotion, and CL is the response/behavior component. Ref. [66] revealed that PV has a strong influence on CS and serves as a critical factor in assessing the degree to which consumers are satisfied. Therefore, we offer the following hypothesis:

H3. 

Perceived value is positively related to customer satisfaction.

2.2.4. CSR and Customer Loyalty

Ref. [67] characterized loyalty as a positive attitude towards a brand that results in consistent purchases of the brand over a long period. CL assesses a customer’s likelihood to repurchase and participate in partnership endeavors [68]. CL can be defined as a single notion, such as an attitude towards the loyalty object or recurring client behavior [69]. Retaining the loyalty of existing consumers is considered a method for maintaining a competitive edge [70]. Customers evaluate a company based on its CSR initiatives, which form their attitudes and behaviors [71]. Ref. [72] conducted a study to explore the effects of CSR on CL in cooperative banks versus commercial banks in the Basque Country via 572 respondents. They implied that CSR is an effective approach and that the social dimension of CSR is the most important element for developing CL. Customers are not only requiring quality products and services, but they also insist that ethical issues be taken into consideration. Ref. [7] proposed that CSR plays the primary role in fostering CL at the expense of company capabilities. CSR efforts may boost the company’s credibility, authenticity, and originality among rivals. Ref. [45] indicated that the integration of CSR practices in hotel management yields multifaceted benefits, as it not only strengthens relationships with customers but also contributes to societal and environmental well-being, ultimately leading to increased CL. Hence, we suggest the following hypothesis:

H4. 

CSR is positively related to customer loyalty.

2.2.5. Perceived Value and Customer Loyalty

According to [73], CL is a highly valuable intangible asset for organizations. Loyalty, in its attitudinal and behavioral aspects, has a significant potential for distinction and is a source of competitive advantage. Prior research found that PV positively influences CL across various industries, including communication, retail, and tourism [9,21,24,74]. Ref. [75] indicated that PV is a major determinant of customer loyalty in the telecommunications sector. Ref. [76] investigate the direct and indirect influence of PV on CL on customer satisfaction from 848 consumer reviews of their online banking experiences. The study found that PV has a significant effect on CL, identifying it as another key driver of CL. Ref. [23] revealed that PV is a strong predictor of CL, exerting a direct influence and demonstrating a greater overall effect on service loyalty than any other construct. Ref. [77] propose that by providing more customer value, service providers can foster trust and commitment, resulting in more loyal customers. Thus, we advocate the following hypothesis:

H5. 

Perceived value is positively related to customer loyalty.

2.2.6. Customer Satisfaction and Customer Loyalty

Satisfaction is the fundamental motivator of the entire quality approach, and CS assessments have been primarily focused on consumers’ perceptions of the quality of products and services [78]. CS is a measurement outcome of a customer’s comparison of desired accomplishments to perceived actual performance [79]. Satisfaction is broadly defined as a post-choice-evaluation judgment [80]. According to [81], CS plays a pivotal role in linking customers’ perceptions of a product’s price and quality with their behavioral intentions and subsequent actions. CS is closely tied to CL, as higher satisfaction levels are indicative of stronger customer engagement and commitment [82].

Previous research indicated that CS has a positive direct and indirect effect on CL in such different contexts as banking [19,83,84,85,86], retailing [87,88,89,90], hospitality [13,55,91,92,93], and telecommunications [7,15,24,94]. Ref. [22] discovered that CS causes higher levels of perceived switching cost and lower levels of perceived alternative attractiveness, hence increasing CL towards mobile internet services. Ref. [87] indicated that the highest level of CS promotes higher CL and repeat purchases over time. In the tourism sector, ref. [6] found that CS has a significant and direct positive effect on CL, positioning it as a key antecedent of loyalty. This study revealed that the higher the customers’ satisfaction with a hotel, the more attitudinally loyal they become, indicating a higher likelihood of repeat patronage and a greater propensity to recommend the hotel to others. According to [95], CS has a considerable influence on CL, with higher satisfaction levels resulting in stronger CL. Hence, we offer the following hypothesis:

H6. 

Customer satisfaction is positively related to customer loyalty.

2.2.7. The Moderating Effect of Demographics

Ref. [96] stated that gender influences the interaction between the fundamentals of the relationship between marketing and CL. Ref. [94] found that age positively affects the link between CS and CL. Similarly, ref. [87] reported that gender moderates the relationship between various dimensions of hedonic shopping values and CS, ultimately influencing loyalty and patronage intentions. Ref. [7] observed that gender has a positive effect on CL, while education level showed no significant impact. Ref. [97], in a study of 487 U.S. hotel customers, found that gender positively moderated the relationship between corporate social responsibility (CSR) and CL. Ref. [98] found that older consumers are more conventional and less likely to try new brands due to limited mobility, which restricts brand choice. Ref. [99] reported that females form stronger brand and interpersonal relationships than males, indicating higher brand loyalty. Ref. [100] highlighted age as a moderating factor between CS and CL, with older consumers exhibiting greater loyalty. Ref. [101] noted that highly educated consumers process information more critically, which can lead to reduced loyalty. Demographic characteristics often serve as significant moderating factors in driving consumer behavior [102,103]. Therefore, we convey the following hypotheses:

H7a. 

Gender moderates the effects of CSR, PV, and CS on CL.

H7b. 

Age moderates the effects of CSR, PV, and CS on CL.

H7c. 

Education background moderates the effects of CSR, PV, and CS on CL.

The conceptual framework is shown in Figure 1.

3. Methodology

Meta-analytic techniques enable the evaluation of several independent or moderator variables using regressions [104]. In this research, the meta-analysis adds additional information, allowing the hypotheses to be tested using moderators that were not addressed in the original inquiry [105].

3.1. Literature Search and Inclusion Criteria

Following one of the best approaches in meta-analytic marketing research [26], we conducted our searches using a variety of catchphrases and combinations of them, including CSR, SQ, PV, CS, CL, gender, age, and educational background. We searched numerous scholarly databases for related publications, including Scopus, Web of Science, ProQuest, JSTOR, ScienceDirect, Emerald, SAGE, Frontiers, ResearchGate, and Springer. Furthermore, to discover any relevant unpublished work and to minimize the file drawer problem, we searched Google and Google Scholar related to marketing and management from 2004 to 2024.

The data started with the aggregate of effect sizes across studies, and each study was required to produce a correlation matrix or other information that could be translated into a correlation coefficient. We used two criteria to determine the number of papers eligible for inclusion. First, primary research ought to be empirical and quantitative. Second, studies are required to report a correlation coefficient (CSR, SQ, PV, CS, and CL) or adequate statistics that would allow for an effect size conversion. Last, a manual search was conducted across leading peer-reviewed journals (e.g., Journal of Business Research, Journal of Retailing and Consumer Services, Journal of Vacation Marketing, Journal of Hospitality Marketing & Management, Journal of Hospitality and Tourism Management, and International Journal of Information Management) and as well as other high-ranking journals in the fields of marketing, business management, and tourism studies. The initial data collection was conducted in September 2024, resulting in a total of 165 published and unpublished studies. From this pool, 11 theoretical articles that did not align with the objectives of the present meta-analysis were excluded. Additionally, 31 studies were removed due to the absence of statistical data or insufficient information necessary for effect size conversion. Following the exclusion process, the final sample for meta-analysis comprised 123 eligible articles. Figure 2 presents the PRISMA flowchart.

3.2. Variable Coding

This study encrypted two independent variables, CS and CL, as well as the aggregate term “published”. The CSR, SQ, and PV were classified as positive work outcomes. According to previous meta-analytic studies [106], if a study provided a multifaceted level effect size for a construct, it was regarded as having an estimated single effect size. In light of the variation in the presentation of demographic information on respondents in each of the selected articles, three broad characteristics of each sample were identified as moderators: gender (male/female), age (younger than 25 years and older), and education background (undergraduate/postgraduate).

3.3. Analytical Techniques

After the coding process, we performed the data analysis. In this study, the correlation coefficients (r) were employed as the primary effect size in the sample, and converted standardized regression coefficients (β) were turned to r for further research. The strength of association between two variables is often interpreted using standardized benchmarks proposed by [107] and supported by [108]. According to these conventions, a correlation coefficient (r) of 0.10 indicates a small effect, 0.30 a medium effect, and 0.50 a large effect. These thresholds offer general guidance for evaluating the magnitude of relationships across studies, providing a consistent framework for interpreting effect sizes within the social and behavioral sciences. Afterward, the study calculated the 95 percent confidence interval (CI) to determine the statistical significance of each impact magnitude. When the 95 percent confidence interval did not include zero, the effect size was deemed to be statistically significant. We adopted [104] Q-statistics to assess the homogeneity of the effect size distribution. The requirement was that the Q-value had to be larger than χ² (χ² with degree of freedom equals (n − 1), where n = number of studies). We followed the work of [108] to identify the possible influences of moderators via significant Q-statistics. This means the observed effect is heterogeneous and that moderators are required to clarify the additional variation in the findings.

The I² statistic quantifies the proportion of variability in effect estimates that is due to heterogeneity rather than chance, providing a measure of dispersion across studies [109]. To assess publication bias, the author references the fail-safe N approaches proposed by [110,111], which are widely employed to estimate the number of unpublished studies needed to nullify the observed effect size. This method assumes that the missing studies have no substantive impact on the overall findings. When the calculated Z-values are positive and the p-value is below 0.05, the author concludes that there is no publication bias. Additionally, we employed the z-test initiated by [112] to evaluate the statistical importance of differences between the groups.

4. Results

Table 1. Meta-analysis results of main effects.

Hyp.	Variables		N	k	Effect Size and 95% Confidence Interval			Heterogeneity				Fail-Safe N
	Independent	Dependent			r	LCI	UCI	p-Value	χ2	Q	I2	Z-Value	Signifcant
H1	CSR	CS	8399	23	0.482	0.465	0.498	0.000	33.92	763.498	97.119	47.944	0.000
H2	SQ	CS	14,697	33	0.540	0.529	0.552	0.000	46.19	2185.02	98.535	73.040	0.000
H3	PV	CS	12,084	29	0.558	0.546	0.570	0.000	41.33	1261.49	97.780	69.005	0.000
H4	CSR	CL	8984	21	0.398	0.380	0.415	0.000	31.41	973.440	97.945	39.734	0.000
H5	PV	CL	4715	14	0.447	0.424	0.470	0.000	22.36	191.017	93.194	32.909	0.000
H6	CS	CL	32,422	78	0.503	0.495	0.511	0.000	98.48	3228.87	97.615	99.351	0.000

Notes: CSR: corporate social responsibility, SQ: service quality, PV: perceived value, CS: customer satisfaction, CL: customer loyalty. Abbreviations: p, level of statistical significance; LCI, lower confidence interval; UCI, upper confidence interval.

states the findings for the main impact. Regarding publication bias, the fail-safe N analysis indicates that all constructs exhibit positive Z-values with p-values below 0.05, suggesting that the observed effect sizes are unlikely to be influenced by a significant publication bias. This study found that CSR is positively related to CS (r = 0.482), and the adjusted 95 percent confidence interval (CI) ranges from 0.465 to 0.498, which does not contain zero, indicating that H1 is supported. The forest plots for Hypothesis 1 are presented in Figure 3.

Furthermore, we noticed that SQ and PV had a strongly positive effect on CS (r = 0.540 and r = 0.558, respectively), in that the corrected 95 percent CI stood from 0.529 to 0.552 for SQ, and the CI varied from 0.546 to 0.570 for PV, with respect to the CIs that exclude zero. Thus, H2 and H3 are supported. Forest plots for Hypotheses 2 and 3 are presented in Figure 4 and Figure 5, respectively.

We also uncovered that CSR is positively related to CL (r = 0.398), and the adjusted 95 percent confidence interval (CI) ranges from 00.380 to 0.415, which does not include zero, implying that H4 is supported. Furthermore, the findings reveal that PV is positively related to CL (r = 0.447), with a corrected 95 percent CI, and ranged from 0.424 to 0.470, which does not comprise zero. Therefore, H5 is supported. Similarly, we found that CS had a strongly positive influence on CL (r = 0.503), and the 95 percent CI fluctuated between 0.495 and 0.511 with a non-zero value. Hence, H6 is supported. Forest plots for Hypotheses 4, 5, and 6 are presented in Figure 6, Figure 7, and Figure 8, respectively.

Furthermore, the impacts of CSR, PV, and CS on CL were confirmed to have significantly larger Q-values than χ². This observation suggests the presence of moderators [108].

Table 2. Meta-analysis results of moderator effects.

Variables		N	k	Effect Size and 95% Confidence Interval			Heterogeneity				Significant Difference
Independent	Dependent			r	LCI	UCI	p- Value	Chi- Square	Q	I- Squared
CSR	CL	8984	21	0.398	0.380	0.415	0.000	31.41	973.440	97.945
Gender
Male		4723	11	0.313	0.287	0.338	0.000	32.67	371.468	97.308	Y
Female		4261	10	0.484	0.461	0.507	0.000	23.68	508.559	98.230
Age
Less than 25		1658	5	0.304	0.260	0.348	0.000	38.88	309.219	98.706	Y
Higher than 25		7326	16	0.418	0.398	0.436	0.000	19.67	641.475	97.662
Education Background
Undergraduate		6373	16	0.426	0.405	0.445	0.000	33.92	910.339	98.352	Y
Postgraduate		2207	5	0.274	0.234	0.312	0.000	26.29	5.197	23.039
PV	CL	4715	14	0.447	0.424	0.470	0.000	22.36	191.017	93.194
Gender
Male		1770	7	0.620	0.590	0.648	0.000	11.07	194.378	96.913	Y
Female		2945	7	0.398	0.367	0.428	0.000	3.841	81.809	92.666
Age
Less than 25		1505	5	0.477	0.437	0.515	0.000	7.814	106.665	96.250	N
Higher than 25		3210	9	0.495	0.468	0.521	0.000	3.841	269.717	97.034
Education Background
Undergraduate		2348	8	0.453	0.420	0.485	0.000	5.991	351.770	98.010	N
Postgraduate		2367	6	0.524	0.494	0.553	0.000	3.841	14.995	66.656
CS	CL	32,422	78	0.503	0.495	0.511	0.000	98.48	3228.87	97.615
Gender
Male		15,548	40	0.475	0.463	0.487	0.000	19.67	1584.44	97.539	Y
Female		16,874	38	0.529	0.518	0.539	0.000	14.06	1603.16	97.692
Age
Less than 25		4707	13	0.549	0.529	0.569	0.000	31.41	373.94	96.791	Y
Higher than 25		27,715	65	0.495	0.486	0.504	0.000	5.991	2832.70	97.741
Education Background
Undergraduate		27,453	64	0.524	0.516	0.533	0.000	28.86	2764.84	97.721	Y
Postgraduate		4969	14	0.376	0.351	0.399	0.000	11.07	317.193	95.902

Notes: Y: yes; N: no, CSR: corporate social responsibility, PV: perceived value, CS: customer satisfaction, CL: customer loyalty. Abbreviations: p, level of statistical significance; LCI, lower confidence interval; UCI, upper confidence interval.

presents the findings of the moderators’ impact on the effect of CSR, PV, and CS on CL. First of all, gender was a significant moderator of all three relationships (CSR-CL, PV-CL, CS-CL). Accordingly, the results indicate a significant gap exists between male and female customers in their need for purchase behavior, with the female group (r = 0.484) having higher correlation scores on the CSR scale than the male group (r = 0.313) in the CSR-CL relationship. However, contradictory results were obtained for the PV-CL relationship, with higher correlations for the male group (r = 0.620) than for the female group (r = 0.398). Additionally, the female group (r = 0.529) has more CS than the male group (r = 0.475) in the case of the CS-CL relationship. Therefore, H7a is totally supported.

Regarding age, Table 2 presents that individuals higher than the age of 25 (r = 0.418) had a higher CSR for CL than those under the age of 25 (r = 0.304). Furthermore, those under 25 (r = 0.549) have a higher CS for CL than those over 25 (r = 0.495). However, no substantial difference was seen in the PV-CL linkages. Hence, H7b is partially supported.

With respect to education background, undergraduates (r = 0.426) have greater CSR toward CL than postgraduates (r = 0.274) in the link of CSR-CL. Similarly, undergraduates (r = 0.524) have stronger CS toward CL than postgraduates (r = 0.376) in the link of CS-CL. In contrast, we discovered no significant variations in the linkages between PV-CL. Thus, H7c is partially supported. The studies included in the meta-analysis are listed in the

Table A1. The studies included in the meta-analysis.

Al-Ghamdi and Badawi [136], 6, (CSR-CS, CSR-CL, CS-CL), 2019

Ahmad et al. [85], 49, (CSR-CS, CS-CL), 2021

Ali et al. [93], 57, (CSR-CS, CS-CL), 2021

Akbari et al. [119], 62, (CSR-CS, CSR-CL, PV-CL, CS-CL), 2020

Agyei et al. [137], 59, (CSR-CS), 2021

Aga & Safakli [138], 53, (SQ-CS), 2007

Ahmed et al. [57], 67, (SQ-CS, CS-CL), 2022

Agarwal & Dhingra [139], 14, (SQ-CS, CS-CL), 2023

Atulkar & Kesari [87], 31, (PV-CS, CS-CL), 2017

Ahmad et al. [140], 57, (CSR-CL), 2021

Aramburu & Pescador [72], 30, (CSR-CL), 2017

Bouichou et al. [141], 12, (CSR-CS), 2022

Babin et al. [142], 33, (SQ-CS), 2005

Bashir et al. [17], 13, (SQ-CS), 2023

Boonlertvanich [19], 17, (SQ-CS, CS-CL), 2018

Bello et al. [51], 62, (SQ-CS), 2020

Cuesta-Valiño et al. [45], 55, (CSR-CS, CSR-CL, CS-CL), 2023

Chang & Yeh [143], 68, (SQ-CS, CS-CL), 2017

Cheng et al. [144], 24, (SQ-CS, CS-CL), 2008

Chiou et al. [145], 50, (SQ-CS, CS-CL), 2020

Chatzoglou et al. [90], 14, (PV-CS, CS-CL), 2022

Chen [146], 69, (PV-CS), 2008

Chen & Tsai [21], 65, (PV-CS, PV-CL, CS-CL), 2008

Chitty et al. [147], 48, (PV-CS, CS-CL), 2007

Chuah et al. [22], 31, (PV-CS, CS-CL), 2017

Correa et al. [148], 57, (PV-CS), 2021

Chen & Cheng [24], 66, (PV-CS, PV-CL, CS-CL), 2012

Chen & Lin [149], 64, (PV-CS), 2015

Chen & Hu [150], 15, (PV-CL), 2010

Chaudhary et al. [151], 20, (PV-CL), 2017

Camilleri & Filieri [152], 15, (CS-CL), 2023

Carmo et al. [89], 44, (CS-CL), 2022

Chen [153], 31, (CS-CL), 2012

Croitoru et al. [95], 10, (CS-CL), 2024

Deng et al. [15], 23, (SQ-CS, CS-CL), 2010

Eggert & Ulaga [154], 36, (PV-CS), 2002

Eid [155], 18, (CS-CL), 2013

El-Adly [6], 31, (CS-CL), 2018

El-Adly & Eid [88], 31, (CS-CL), 2016

Gong & Yi [118], 51, (SQ-CS, CS-CL), 2018

Gli et al. [156], 68, (CSR-CL), 2024

Gürlek et al. [157], 62, (CSR-CL), 2017

Ganguli & Roy [84], 17, (CS-CL), 2011

Hassan et al. [158], 54, (CSR-CS), 2022

Hsu [159], 71, (CSR-CS, CSR-CL, CS-CL), 2018

Hu et al. [160], 70, (SQ-CS, PV-CS), 2009

Hussain et al. [161], 43, (SQ-CS, PV-CS), 2015

Hwang et al. [16], 31, (SQ-CS), 2021

Han & Hyun [30], 15, (SQ-CS), 2017

Hafidz & Huriyahnuryi [162], 25, (PV-CS, CS-CL), 2023

Hang & Trung [163], 6, (CS-CL), 2024

Hau & Thuy [164], 60, (CS-CL), 2012

Islam et al. [7], 58, (CSR-CS, CSR-CL, CS-CL), 2021

Jiang et al. [165], 60, (PV-CL), 2016

Jain et al. [92], 19, (CS-CL), 2019

Kim & Kim [97], 3, (CSR-CS, CS-CL), 2016

Klaysung et al. [14], 26, (CSR-CS, CS-CL), 2022

Khoo [166], 52, (SQ-CS), 2020

Kim [55], 40, (SQ-CS, CS-CL), 2011

Kuo et al. [64], 8, (SQ-CS, PV-CS), 2009

Al-Ghamdi and Badawi [136], 6, (CSR-CS, CSR-CL, CS-CL), 2019

Ahmad et al. [85], 49, (CSR-CS, CS-CL), 2021

Ali et al. [93], 57, (CSR-CS, CS-CL), 2021

Akbari et al. [119], 62, (CSR-CS, CSR-CL, PV-CL, CS-CL), 2020

Agyei et al. [137], 59, (CSR-CS), 2021

Aga & Safakli [138], 53, (SQ-CS), 2007

Ahmed et al. [57], 67, (SQ-CS, CS-CL), 2022

Agarwal & Dhingra [139], 14, (SQ-CS, CS-CL), 2023

Atulkar & Kesari [87], 31, (PV-CS, CS-CL), 2017

Ahmad et al. [140], 57, (CSR-CL), 2021

Aramburu & Pescador [72], 30, (CSR-CL), 2017

Bouichou et al. [141], 12, (CSR-CS), 2022

Babin et al. [142], 33, (SQ-CS), 2005

Bashir et al. [17], 13, (SQ-CS), 2023

Boonlertvanich [19], 17, (SQ-CS, CS-CL), 2018

Bello et al. [51], 62, (SQ-CS), 2020

Cuesta-Valiño et al. [45], 55, (CSR-CS, CSR-CL, CS-CL), 2023

Chang & Yeh [143], 68, (SQ-CS, CS-CL), 2017

Cheng et al. [144], 24, (SQ-CS, CS-CL), 2008

Chiou et al. [145], 50, (SQ-CS, CS-CL), 2020

Chatzoglou et al. [90], 14, (PV-CS, CS-CL), 2022

Chen [146], 69, (PV-CS), 2008

Chen & Tsai [21], 65, (PV-CS, PV-CL, CS-CL), 2008

Chitty et al. [147], 48, (PV-CS, CS-CL), 2007

Chuah et al. [22], 31, (PV-CS, CS-CL), 2017

Correa et al. [148], 57, (PV-CS), 2021

Chen & Cheng [24], 66, (PV-CS, PV-CL, CS-CL), 2012

Chen & Lin [149], 64, (PV-CS), 2015

Chen & Hu [150], 15, (PV-CL), 2010

Chaudhary et al. [151], 20, (PV-CL), 2017

Camilleri & Filieri [152], 15, (CS-CL), 2023

Carmo et al. [89], 44, (CS-CL), 2022

Chen [153], 31, (CS-CL), 2012

Croitoru et al. [95], 10, (CS-CL), 2024

Deng et al. [15], 23, (SQ-CS, CS-CL), 2010

Eggert & Ulaga [154], 36, (PV-CS), 2002

Eid [155], 18, (CS-CL), 2013

El-Adly [6], 31, (CS-CL), 2018

El-Adly & Eid [88], 31, (CS-CL), 2016

Gong & Yi [118], 51, (SQ-CS, CS-CL), 2018

Gli et al. [156], 68, (CSR-CL), 2024

Gürlek et al. [157], 62, (CSR-CL), 2017

Ganguli & Roy [84], 17, (CS-CL), 2011

Hassan et al. [158], 54, (CSR-CS), 2022

Hsu [159], 71, (CSR-CS, CSR-CL, CS-CL), 2018

Hu et al. [160], 70, (SQ-CS, PV-CS), 2009

Hussain et al. [161], 43, (SQ-CS, PV-CS), 2015

Hwang et al. [16], 31, (SQ-CS), 2021

Han & Hyun [30], 15, (SQ-CS), 2017

Hafidz & Huriyahnuryi [162], 25, (PV-CS, CS-CL), 2023

Hang & Trung [163], 6, (CS-CL), 2024

Hau & Thuy [164], 60, (CS-CL), 2012

Islam et al. [7], 58, (CSR-CS, CSR-CL, CS-CL), 2021

Jiang et al. [165], 60, (PV-CL), 2016

Jain et al. [92], 19, (CS-CL), 2019

Kim & Kim [97], 3, (CSR-CS, CS-CL), 2016

Klaysung et al. [14], 26, (CSR-CS, CS-CL), 2022

Khoo [166], 52, (SQ-CS), 2020

Kim [55], 40, (SQ-CS, CS-CL), 2011

Kuo et al. [64], 8, (SQ-CS, PV-CS), 2009

Kim et al. [167], 34, (PV-CS), 2020

Kim & Kim [97], 63, (CSR-CL), 2016

Kao et al. [168], 1, (CS-CL), 2008

Kondasani & Panda [169], 27, (CS-CL), 2015

Lee [12], 44, (CSR-CS, CSR-CL, CS-CL), 2018

Latif et al. [13], 15, (SQ-CS, CSR-CL, CS-CL), 2020

Lee et al. [56], 41, (SQ-CS), 2020

Liu et al. [91], 45, (SQ-CS, CS-CL), 2016

Lai et al. [23], 29, (PV-CS, PV-CL, CS-CL), 2009

Lee & Moghavvemi [65], 4, (PV-CS, CS-CL), 2015

Lin & Wang [170], 28, (PV-CS, PV-CL, CS-CL), 2006

Liu et al. [66], 6, (PV-CS), 2024

Lu & Hsiao [171], 28, (PV-CS), 2010

Lam et al. [172], 32, (PV-CS, CS-CL), 2004

Le [173], 21, (CSR-CL), 2022

Leclercq-Machado et al. [86], 57, (CSR-CL, CS-CL), 2022

Li & Li [174], 57, (CSR-CL), 2024

Lei et al. [175], 12, (CS-CL), 2022

Loureiro [176], 39, (CS-CL), 2010

Muflih [8], 31, (CSR-CS, CSR-CL), 2021

Ma [177], 74, (SQ-CS), 2021

Mukerjee [74], 35, (PV-CL), 2018

Mahmood et al. [178], 73, (CS-CL), 2018

Makanyeza & Chikazhe [179], 17, (CS-CL), 2017

Malki et al. [180], 61, (CS-CL), 2024

Manyanga et al. [100], 6, (CS-CL), 2022

Molinillo et al. [10], 31, (CS-CL), 2022

Nguyen et al. [181], 57, (SQ-CS, PV-CS, CS-CL), 2018

Nyadzayo & Khajehzadeh [77], 31, (PV-CL, CS-CL), 2016

Nobar & Rostamzadeh [182], 47, (CS-CL), 2016

Opata et al. [183], 66, (CS-CL), 2019

Orazgaliyeva et al. [184], 56, (CS-CL), 2024

Park et al. [5], 29, (CSR-CS, CS-CL), 2017

Paulose & Shakeel [9], 42, (PV-CS, PV-CL, CS-CL), 2021

Rivera et al. [185], 61, (CSR-CS), 2016

Rajic et al. [186], 11, (SQ-CS, PV-CS), 2013

Ryu et al. [187], 16, (PV-CS), 2012

Raza et al. [188], 57, (CSR-CL), 2020

Shahzad et al. [189], 5, (CSR-CS), 2019

Sindhu and Arif [190], 6, (CSR-CS, CSR-CL, CS-CL), 2017

Satti et al. [59], 2, (SQ-CS, CS-CL), 2020

Simiyu & Bonuke [53], 22, (SQ-CS, CS-CL), 2023

Suhartanto et al. [191], 38, (SQ-CS, CS-CL), 2019

Servera-Francés & Piqueras-Tomás [192], 9, (PV-CS, PV-CL, CS-CL), 2019

Suhartanto et al. [193], 45, (PV-CS, PV-CL, CS-CL), 2018

Shah & Jan [194], 59, (CSR-CL), 2021

Sánchez-Fernández et al. [195], 16, (CS-CL), 2020

Santouridis & Trivellas [94], 67, (CS-CL), 2010

Schirmer et al. [196], 37, (CS-CL), 2016

Serra-Cantallops et al. [197], 61, (CS-CL), 2018

Tran [198], 6, (CSR-CS, CSR-CL, CS-CL), 2022

Tran & Nguyen [199], 57, (CSR-CS), 2020

Tuncer et al. [58], 42, (SQ-CS, PV-CS), 2020

Thuy & Hau [200], 17, (CS-CL), 2010

Uzir et al. [25], 31, (SQ-CS, PV-CS), 2021

Woratschek et al. [201], 70, (CS-CL), 2019

Yuen et al. [44], 66, (CSR-CS, SQ-CS, CS-CL), 2016

Yang & Peterson [76], 51, (PV-CL, CS-CL), 2005

Yum & Yoo [202], 57, (CS-CL), 2023

Yusra & Agus [203], 72, (CS-CL), 2018

Zygiaris et al. [204], 12, (SQ-CS), 2022

Kumar & Lata [205], 46, (SQ-CS), 2021

Notes: Codes in parentheses: CSR: corporate social responsibility, SQ: service quality, PV: perceived value, CS: customer satisfaction, CL: customer loyalty. Journals are footnoted in alphabetical order: (1) Asia Pacific Journal of Tourism Research; (2) Asia Pacific Journal of Innovation and Entrepreneurship; (3) Asian Journal of Sustainability and Social Responsibility; (4) Asian Journal of Business and Accounting; (5) Business & Economic Review; (6) Cogent Business & Management; (7) Corporate Reputation Review; (8) Computers in Human Behavior; (9) Economic Research-Ekonomska Istraživanja; (10) European Research on Management and Business Economics; (11) E+M Ekonomie a Management; (12) Frontiers in Psychology; (13) FinTech; (14) Heliyon; (15) International Journal of Hospitality Management; (16) International Journal of Contemporary Hospitality Management; (17) International Journal of Bank Marketing; (18) International Journal of Tourism Research; (19) International Journal of Advanced Science and Technology; (20) International Journal of Management and Commerce Innovations; (21) International Journal of Emerging Markets; (22) International Journal of Financial Services Management; (23) International Journal of Information Management; (24) International Journal of E-Business Research; (25) International Journal of Social Science Education Communication and Economics; (26) International Journal of Health Sciences; (27) International Journal of Health Care Quality Assurance; (28) Information & Management; (29) Journal of Business Research; (30) Journal of Business Ethics; (31) Journal of Retailing and Consumer Services; (32) Journal of the Academy of Marketing Science; (33) Journal of Services Marketing; (34) Journal of Vacation Marketing; (35) Journal of Financial Services Marketing; (36) Journal of Business & Industrial Marketing; (37) Journal of Strategic Marketing; (38) Journal of Islamic Marketing; (39) Journal of Travel & Tourism Marketing; (40) Journal of Hospitality Marketing & Management; (41) Journal of Hospitality and Tourism Management; (42) Journal of Quality Assurance in Hospitality & Tourism; (43) Journal of Air Transport Management; (44) Journal of Promotion Management; (45) Journal of Foodservice Business Research; (46) Journal of Operations and Strategic Planning; (47) Journal of Business Economics and Management; (48) Marketing Intelligence & Planning; (49) Market Forces; (50) Maritime Policy & Management; (51) Psychology and Marketing; (52) PSU Research Review; (53) Problems and Perspectives in Management; (54) Pakistan Journal of Commerce and Social Sciences; (55) Quality & Quantity; (56) Regional Science Policy & Practice; (57) Sustainability; (58) Sustainable Production and Consumption; (59) SAGE Open; (60) Service Business; (61) Spanish Journal of Marketing - EISC; (62) Social Responsibility Journal; (63) SpringerPlus; (64) Technological Forecasting & Social Change; (65) Tourism Management; (66) Total Quality Management & Business Excellence; (67) The TQM Journal; (68) Transport Policy; (69) Transportation Research Part A: Policy and Practice; (70) The Service Industries Journal; (71) The International Journal of Social Sciences and Humanities Invention; (72) The Journal of Social Sciences Research; (73) The Lahore Journal of Business; (74) Telematics and Informatics.

.

5. Discussion

Several discussions can be derived from the findings of this study. First, CSR has a positive effect on CS, which is in line with [5]. We confirmed the findings of [113], according to which CSR activity promotes CS and positive brand purchase intention. CSR fosters CS by appealing to customers’ psychosocial needs and seeking to satisfy their moral and social obligations [44]. We also confirmed the findings of [7], according to which a CSR strategy generates distinct advantages in the market, pleases customers, builds a favorable impression of the company in the eyes of the public, and gives it a platform in which to showcase its commitments, promises, and community-related activities in a way that goes beyond the company’s legal obligations.

Second, this study found that SQ and PV have a strongly positive effect on CS. We verified the findings of [114], according to which satisfaction is an antecedent of SQ and that SQ is a predictor of satisfaction; the two concepts are interchangeable. If a provider’s SQ is higher, customers will be more satisfied, resulting in higher CL [15]. We concurred with [58] that SQ plays a crucial role in shaping CS and that raising SQ and CS leads to customers’ positive behavioral intentions. Additionally, we confirmed the findings of [23], according to which PV can be crucial in helping service firms achieve their objective. When consumers feel that they are receiving branded products at reduced prices and enjoy the satisfaction of haggling, they will view the purchase as having greater value [87]. PV aspects not only guide effective activities but also allow the deployment of limited firm resources to optimal solutions that promote CS [22]. We also validated the work of [66], according to which PV is vital in determining CS and the competitiveness of a product or service.

Third, the findings indicate that CSR has a significant influence on CL. CSR is a vital component of corporate operations [115]. We confirmed the findings of [72], according to which CSR is a good strategy, its social component is the most important for building CL, and CSR may influence customer outcomes through different mediators, such as by boosting brand equity and creating higher levels of identification with the company. The findings also agreed with the work of [86], according to which CSR is a significant aspect for customers, since it fosters their satisfaction and loyalty. Furthermore, this study indicated that PV has a considerable effect on CL and that strong customer-perceived value can considerably improve CL [116]. PV is another important driver of CL, such that a higher possibility of PV will lead to greater CL [21,76].

Next, CS has a strong and positive effect on CL. CS is a crucial requirement for expanding the number of loyal customers. CS is vital for any firm’s growth, as satisfied customers create positive word of mouth about the enterprise, reinforcing the consumers’ intention to return [57,117]. CS is widely recognized as a key predictor of CL [7,24,88,118,119]. We validated the findings of [120], according to which satisfied customers possess a stronger inclination to present loyalty, which may be assessed from a behavioral or attitudinal standpoint.

Last, regarding moderating effects, the results indicate that gender fully moderates the relationships among CSR, PV, CS, and CL, while age and education background serve as partial moderators. These findings align with [96], who observed that female customers exhibit higher loyalty in the service sector. This may be attributed to their greater sociability and stronger inclination to cultivate personal connections and social bonds [87]. We also confirmed [121] age-related finding that older customers are more loyal than younger ones.

6. Practical and Theoretical Implications

This research offers several practical implications. First, we verified the critical role of CSR in determining CL via CS. We propose that CSR can assist in promoting CL by creating emotional connections, strengthening brand image, consumer emotions, and motivations, improving brand image and reputation, delivering value and quality, and increasing CS. Additionally, we expanded upon the work of [7] by assessing the mechanism through which CSR impacts CL in three phases: (1) aligning brand value: CSR creates strong emotional ties with customers through transparency and sincerity regarding social and environmental concerns; (2) enhancing customer experience: CSR programs involve efforts and commitments to impact social and other stakeholders in a clear, consistent, and engaging way that causes customers to feel a positive impact through their purchases (meeting or exceeding customer expectations); (3) creating a sense of community: CSR strengthens relationships between the company and its customers by developing CSR campaigns, events, or programs involving customers (rewards, discounts, coupons, or loyalty points) that contribute to a sense of belonging and engagement. When firms engage in CSR practices such as environmental sustainability, ethical labor standards, and community engagement, they signal integrity and long-term commitment, which in turn strengthen customer trust and enhance brand perception. CSR also supports long-term business success by generating sustainable value for society and the environment. Moreover, prioritizing CSR enables firms to adapt to evolving market dynamics, attract and retain talent, and cultivate strong stakeholder relationships, all of which contribute to a sustained competitive edge and long-term success. We also confirmed the work of [122], according to which CSR plays a vital role in the setting of a sustainable economy, which is familiar with ethical company operations, and a higher level of CSR will yield beneficial behavioral outputs.

Second, we validated the findings of [19,51], confirming that SQ plays a pivotal role in forming CS, which in turn fosters CL. Attributes of SQ, such as outcome quality, physical environment quality, and interactional quality, are essential drivers of CS in the establishment of loyalty [123]. However, according to the SQ model, to boost service quality, the organization must meet indicators for reliability, responsiveness, assurance, empathy, and tangible outcomes [124]. SQ often requires time to effectively influence CL, as loyalty typically results from cumulative and consistent positive experiences rather than isolated interactions. Sustained exposure to high SQ fosters trust, satisfaction, and emotional connection, which serve as key antecedents of CL [47]. For instance, in the era of Industry 4.0, with increasing digitization and AI integration in customer interfaces, SQ still takes time to impact CL. The key challenge is not just delivering high-quality service but sustaining it consistently to build trust and strengthen long-term customer relationships.

Third, we confirmed with the work of [9] that PV is a critical determinant of CL. This study argues that CL is shaped by consumer expectations and experiences, which are influenced by brand image, perceived quality, and innovative features that collectively contribute to the long-term sustainability of a firm. For example, Apple has been among the world’s firms with the highest market value since 2010. Apple’s customers are willing to pay a premium price for its products since Apple provides a high level of quality (i.e., high-tech, simplicity, minimalism, and long-lasting), innovative features (i.e., style, privacy, security, and software updates), reliability (i.e., resale value), and customer service (i.e., the Apple ecosystem), which has resulted in a high level of CS and shapes CL.

Fourth, we confirmed the findings of [6,95], according to which CS is a critical driver of CL, with higher satisfaction levels directly linked to increased loyalty. This study further proposes that satisfied consumers are more likely to become loyal advocates who actively recommend the product or service to others, ultimately enhancing sales performance and contributing to long-term profitability. We agree with [125] that CL effects exist in both the short term, since loyal customers tend to spend more frequently, and in the long term, since firms gain new customers due to their customers’ advocacy and favorable comments. Furthermore, we propose that CL is critical to a company’s long-term success. By stimulating repeat purchases and transforming walk-in customers into loyal repeat customers, enterprises can significantly enhance both revenue and profitability.

Next, the findings indicate a gender discrepancy in CSR, CS, and CL. In particular, this study reveals that female customers tend to support or are more sensitive to CSR than male customers. We confirm with [99] that female customers are more loyal than male customers. An explanation for this is that women prefer to focus on building and sustaining intimate relationships with specific individuals, whilst men tend to focus on establishing and keeping ties with more abstract and larger groups of people [126]. We extend the work of [98] by proposing that older consumers exhibit greater brand loyalty and are more cautious in adopting new products, often valuing tradition and stability. In contrast, younger consumers are more inclined toward novelty, innovation, and new experiences. This study also extends the work of [101] by suggesting that individuals with higher education levels demonstrate lower loyalty, as their purchasing decisions are driven more by rational criteria, such as price, convenience, durability, and practicality, than by emotional attachment.

This study adopted social change and expectancy–value theory to analyze the role of CSR, SQ, PV, and CS in forming CL by employing a meta-analysis. We confirm the work of [35] that as the degree of social exchange intensifies, the perception of shared responsibility between the customer and the organization correspondingly strengthens. This heightened sense of mutual accountability reshapes the emotional dynamics within the exchange relationship. These emotional connections are particularly significant, as prior research has demonstrated their strong influence on customer loyalty toward the service provider [127,128]. Consistent with [26], we acknowledge that a meta-analysis is particularly robust when conclusions are drawn from a large number of studies and participants. By aggregating data across studies, meta-analysis addresses the limitations associated with small sample sizes and enhances the statistical significance of findings [129]. Moreover, it offers greater statistical power than individual studies due to an increased cumulative sample size [130]. As noted by [131], a meta-analysis enables the synthesis of findings from multiple sources, yielding statistically reliable estimates of the strength and direction of relationships between variables.

Additionally, meta-analysts identify relevant moderating variables to account for variations across studies and to resolve conflicting findings in a literature stream [130,132]. This study adds new insights to the literature on CL by unraveling the moderating role of demographic characteristics on the impact of CSR, PV, and CS on CL. Responding to the calls of previous scholars [5,30], we studied the effect of gender, age, and education background on the links among CSR, PV, CS, and CL through the lens of a meta-analysis to reconcile inconsistent findings in prior research and offer clearer guidance for both academics and practitioners [133].

This study has several limitations, indicating suggestions for future research. First, the results of the meta-analysis may be constrained by the selection of an incomplete set of research, the inclusion of studies lacking internal, external, concept, and statistical conclusion validity, and the presence of studies with small sample sizes. Thus, results from heterogeneity approaches make the reasoning for an integrated outcome more difficult [134,135]. Second, our findings were limited to the journals that fulfilled the selection criterion. Third, although we attempted to incorporate all key publications relating to demographic moderating effects, there was a shortage of relevant articles utilizing a meta-analysis approach to corroborate these associations because there are few such research papers on it. Future research can boost the findings by integrating more meta-analysis papers or investigating different types of moderators (income, work experience, family background, etc.) on the relationships between CSR, PV, CS, or other antecedents and CL. Future scholars could determine the mediating impact of other variables on the linkage between CSR and loyalty, such as positive word of mouth, trust credibility, brand image, and affective evaluation [13,56].

Furthermore, to address the limitations related to industry-specific and geographic generalizability in meta-analytic studies, future meta-analyses could strive for more inclusive data collection by incorporating studies from a wider range of geographic locations and industry sectors to enhance the external validity of findings. Researchers are also encouraged to conduct sensitivity analyses to assess the robustness of results across different contexts. The transparent reporting of study characteristics such as country of origin, economic development status, and industry type will further support efforts to evaluate and improve generalizability across diverse settings.

7. Conclusions

This study explored the role of CSR, SQ, and PV in shaping CL via CS by adopting a meta-analytic approach. The findings revealed that CSR enhances CL via CS, underscoring its role in fostering long-term business success through positive and sustainable contributions to society and the environment. CSR can boost CL through three key mechanisms: aligning brand value, enhancing customer experience, and creating a sense of community. Moreover, PV plays a crucial role in shaping CL and sustaining firm performance, as it influences customer expectations and experiences through brand image, perceived quality, and innovative features. CS is a key driver of CL, as satisfied customers are more likely to remain loyal. For that reason, managers and policymakers should prioritize employee training to ensure staff are knowledgeable, empathetic, and responsive to customer needs, ultimately fostering deeply satisfied and loyal customers. Specifically, it is essential to offer personalized experiences that reflect customers’ preferences and desires. This can be achieved by using customer surveys or feedback forms to gather insights into customer views and experiences with the products and services provided. Regarding moderating effects, the findings indicate that female customers are more influenced by CSR initiatives than males. Older customers tend to show higher loyalty, potentially due to a more cautious approach to new products. Individuals with higher education levels exhibit lower loyalty, as they evaluate products based on a broader range of criteria, including price, convenience, durability, and practicality.